Smith v. Texas
543 U.S. ___, 125 S.Ct. 400 (2004) (per curiam)

The Supreme Court, in a 7-2 per curiam opinion, reversed the Texas Court of Criminal Appeals, holding that there was a reasonable likelihood that the jury, which was given a post-Penry I and pre-Penry II sentencing instruction, failed to give effect to relevant mitigating evidence regarding: (1) defendant's diagnosis of potential organic learning disabilities and speech handicaps at an early age; (2) his verbal IQ score of 75 and a full scale IQ of 78, which resulted in defendant's placement in special education classes throughout the bulk of his time in school; (3) defendant's frequently exemplary behavior in school despite his problems; (4) that defendant's father was a drug addict, who was involved in gang violence and other criminal activity, and regularly stole from family members in order to support his drug habit; and (5) defendant's age of 19 at the time of the capital offense.

Petitioner LaRoyce Lathair Smith was charged with capital murder in Texas. The statute in effect at the time of his trial required the jury to determine the sentence based on its answer to two special issues concerning deliberateness of the crime, and future dangerousness. An affirmative answer to both questions required the jury to return a death sentence; a negative answer required a life sentence. This statute was amended sometime after Smith's trial.

Although the sentencing scheme in existence during Smith's trial withstood general constitutional scrutiny in Jurek v. Texas, 428 U.S. 262 (1976), it was found constitutionally inadequate in Penry v. Lynaugh, 492 U.S. 302 (1989) (Penry I), because it failed to provide a means for the jury to give full effect to the defendant's mitigating evidence of mental retardation and childhood abuse. In an effort to avoid error under Penry I, the trial judge in Smith's case gave a supplemental instruction that directed the jury to give effect to the mitigating evidence Smith presented by negating what would otherwise be a positive response to one of the special issues. In pertinent part, the instruction stated:

Ex Parte Smith, 132 S.W..3d 407, 409 (Tex. Crim. App. 2004). During closing arguments at the sentencing phase, the prosecutor reminded the jurors of their duty to answer truthfully the two special issues of deliberateness and future dangerousness. The jury verdict form conformed to the prosecutor's remarks and made no mention of either nullification or mitigation. The jury answered both questions affirmatively and Smith was sentenced to death.

On appeal, he argued that the supplemental instruction failed to give the jury a genuine mechanism for giving his mitigating evidence effect. The Texas Court of Criminal Appeals affirmed the sentence. After the Supreme Court issued Penry v. Johnson, 532 U.S. 782 (2001) (Penry II), Smith again challenged the instruction in state court. In Penry II, the Supreme Court had ruled that a similar nullification instruction was inadequate to ensure that the jury could give effect to the defendant's mitigating evidence. The Texas Court of Criminal Appeals again rejected Smith's claim of constitutional error.

Shortly thereafter, the Supreme Court issued its decision in Tennard v. Dretke, 124 S.Ct. 2562 (2004). Tennard involved a pre-Penry I trial. Both the state and lower federal courts had rejected Tennard's Penry claim on the ground that he had failed to pass a threshold standard for constitutionally relevant evidence, that is, evidence of a uniquely severe permanent handicap with which the defendant was burdened through no fault of his own, and evidence that the capital crime was attributable to this condition.

In Tennard, the Supreme Court rejected this threshold test, finding that it had no support in Supreme Court precedent. In Smith's case, the Texas Court of Criminal Appeals had relied on precisely the same screening test as found constitutionally inadequate in Tennard. The state court had concluded that Smith's low IQ and placement in special education classes was irrelevant because this did not establish a "severe disability." Looking to Tennard, as well as Wiggins v. Smith, 539 U.S. 510, 535 (2003), the Supreme Court responded:

Smith, at 405.

The state court also found Smith's evidence to be irrelevant because he failed to link his troubled childhood or limited mental abilities to the capital offense. The Supreme Court pointed out it had also rejected in Tennard the requirement that a defendant show a "nexus" between the proffered defense evidence and the crime in order for the evidence to be considered "mitigating" under the Eighth Amendment. That Smith's evidence "was relevant for mitigation purposes" was plain even under pre-Tennard precedents. The Texas Court of Criminal Appeals erroneously relied on a test that the Supreme Court "never countenanced" and "now have unequivocally rejected." Id.

Having found that the trial court was required to provide the jury with a means for considering Smith's mitigating evidence, the Supreme Court turned to the "nullification instruction." The Texas Court of Criminal Appeals had ruled in the alternative that even if Smith had proffered relevant mitigating evidence, the special instruction allowed the jury to give effect to his evidence. The Supreme Court disagreed, finding that the instruction at issue was constitutionally indistinguishable from the one condemned in Penry II in that it required the jury to "nullify special issues contained within a verdict form." Id. at 406. It was true that in Penry II the jury was told merely that it "should" answer a special issue negatively if it found a life verdict appropriate in light of the mitigating evidence, whereas Smith's jury was told it "shall" do so under the same circumstances. The Supreme Court concluded that this was an insufficient basis to take Smith's case outside of Penry II. Just as in Penry II, the instructions provided to Smith's jury made it "logically and ethically impossible" for the jury to follow both the standard instructions and the special instruction. Because there was "no principled distinction, for Eighth Amendment purposes, between the instructions given to petitioner's jury and the instructions in Penry II," the judgment of the Texas Court of Criminal Appeals was reversed.

Justice Scalia, joined by Justice Thomas, dissented. Both would affirm. See Walton v. Arizona, 497 U.S. 639, 673 (Scalia, J., concurring in part and concurring in the judgment).

Florida v. Nixon
543 U.S. ___, 125 S.Ct. 551 (2004)

In this death penalty case from Florida, the Supreme Court unanimously (with the Chief Justice absent) reversed the Florida Supreme Court, and held that where a defense attorney concedes the defendant’s guilt in the first phase of a bifurcated capital trial, and the defendant was aware of counsel’s intentions but neither consented nor objected to the strategy, the reasonableness of counsel’s actions must be judged under the standard of Strickland v. Washington.

Defendant Joe Nixon confessed to kidnaping and killing a woman who had given him a ride after he approached her in a mall asking for assistance in jump-starting his car. He tied the woman to a tree in a remote area and set fire to her and some of her belongings before fleeing in her car.

The prosecution gathered what was described as overwhelming evidence corroborating Nixon’s confession. The assistant public defender assigned to represent Nixon filed a plea of not guilty and deposed all of the State’s witnesses. After determining that Nixon’s guilt was not subject to reasonable dispute, he commenced plea negotiations hoping to persuade the prosecutor to accept a guilty plea to all charges in exchange for dropping the death penalty. The State, however, was unwilling to recommend a sentence less than death.

At this juncture, defense counsel, who was experienced in capital cases, determined that the sole hope for obtaining a life sentence for Nixon was to present extensive mitigating evidence concerning Nixon’s mental instability. He believed that by denying guilt at the first phase of the trial, his ability to persuade the jury that the crime was attributable to Nixon’s mental illness would be compromised. Counsel therefore concluded that the best strategy would be a concession of guilt which would preserve his credibility at the penalty phase. Nixon’s attorney attempted to explain this strategy to Nixon on a number of occasions, but Nixon was unresponsive. He never verbally approved or disapproved of his attorney’s tactical decision, and he gave little, if any, assistance or direction in the case preparation.

During jury selection, Nixon began exhibiting bizarre and violent behavior, leading to a waiver of his right to attend trial. Nixon was not present when his attorney gave an opening statement that conceded Nixon’s guilt, and told the jury that the case was about the appropriate penalty. During the guilt phase, defense counsel did cross-examine witnesses when he thought the testimony needed clarification and successfully objected to the admission of prejudicial evidence. He did not present a case in defense. In his closing argument, he again conceded guilt, and urged the importance of the upcoming penalty phase. After Nixon was convicted, defense counsel presented mitigating evidence concerning Nixon’s mental and emotional problems. The jury nevertheless recommended death, and the trial court imposed that punishment.

In post-conviction proceedings, new counsel litigated the question of whether trial counsel performed ineffectively by conceding Nixon’s guilt without obtaining his express consent. The Florida Supreme Court ruled that a concession of guilt by counsel requires an affirmative, explicit acceptance by the defendant, without which trial counsel’s performance is presumptively inadequate. The court reasoned that such an action by counsel is tantamount to a guilty plea. Because the record failed to show an express acceptance to the strategy by Nixon, the Florida Supreme Court ordered a new trial. In so doing, it applied a presumption of prejudice under United States v. Cronic, 466 U.S. 648 (1984), which is reserved for situations where counsel completely failed to function as the client’s advocate.

The Supreme Court, in an opinion by Justice Ginsburg, began by acknowledging that there are a few decisions that a defendant has ultimate authority over: whether to plead guilty, waive a jury, testify, or take an appeal. The Court disagreed, however, with the state court’s conclusion that trial counsel’s concession in this case was the functional equivalent of a guilty plea. It pointed out that the prosecution retained the burden of providing the jury with competent evidence establishing all elements of the charged crimes beyond a reasonable doubt. Further, defense counsel retained the ability to cross-examine witnesses, and to exclude prejudicial evidence. And Nixon could appeal any errors occurring during the trial or in the instructions. Thus, defense counsel’s concession to the jury was not, in the Court’s view, fairly analogized to a plea of guilty.

The Court did find that trial counsel was required to discuss the proposed strategy with his client before making the concession. Here, the record established that counsel did so. Given Nixon’s resistance to answering inquiries from both counsel and the trial court, the Supreme Court concluded that trial counsel was “not additionally required to gain express consent before conceding Nixon’s guilt.” Nixon at 561.

Because the Florida Supreme Court had erroneously equated counsel’s concession with a guilty plea, that led it to mistakenly apply Cronic. In the Supreme Court’s view, counsel’s action here did not rank as a failure to function as an adversary in any meaningful way. Thus, the appropriate test to be applied in this case is that of Strickland v. Washington, 466 U.S. 668 (1984), and the correct question is whether counsel’s decision to concede guilt was objectively unreasonable.

In determining that trial counsel’s performance was not necessarily deficient, the Supreme Court looked to the ABA death penalty guidelines, as well as law review articles discussing the relationship between guilt phase defenses and the penalty phase. The Court also noted that Clarence Darrow adopted a strategy similar to Nixon’s lawyer when representing Loeb and Leopold. The Court concluded:

Nixon at 563.

Brown v. Payton
544 U.S. ___, 125 S.Ct. 1432 (2005)

In a 5-3 decision, with the Chief Justice not participating, the Supreme Court held in this California death penalty case that 28 U.S.C. § 2254(d) precluded habeas relief on a claim that there was a reasonable likelihood the jury believed it was precluded from considering evidence of Payton's post-crime religious conversion in deciding his sentence. Justice Kennedy, joined by Justices O'Connor, Scalia, Thomas and Breyer, delivered the opinion of the Court.

William Payton was convicted in California of a rape-murder and two attempted murders. The victims were all stabbed by Payton. At the sentencing phase, the prosecutor presented evidence that Payton had previously stabbed a girlfriend, that he had prior convictions for rape and a drug-related felony, and that he had admitted while in jail that he had an "urge to kill," as well as problems with sex and women that caused him to view all women as persons to stab and rape. The case in mitigation consisted of witnesses who described Payton's post-crime religious conversion and the calming effect he had on other prisoners.

The jury was instructed to consider 11 different factors in determining Payton's sentence. The final factor, factor (k), is a catch-all mitigator: "Any other circumstance which extenuates the gravity of the crime even though it is not a legal excuse for the crime." Defense counsel unsuccessfully argued that the language of factor (k) should be modified to clarify for the jury that it encompassed evidence about Payton's background and character. Although the trial court agreed that factor (k) was intended to be that broad, contrary to the prosecutor's narrow interpretation of the scope of the factor, it left it to defense counsel to argue to the jury the evidence that fell within it.

Prior to closing arguments, the trial judge instructed the jury that counsel was not providing evidence, and that it should rely on its own recollection of the evidence. The prosecutor then argued to the jury that factor (k) did not permit the jury to consider anything that occurred after the capital crime. A defense objection and request for a mistrial was denied. The trial judge did admonish the jury that the prosecutor's comments were merely argument, but it did not tell the jury that the prosecutor's interpretation of factor (k) was incorrect. Although the prosecutor repeated several times his view that no mitigating evidence had been presented by Payton, he also discussed the defense evidence in detail and contended that the aggravating facts outweighed Payton's religious conversion. Defense counsel countered by asserting that factor (k) was designed to cover exactly the type of evidence that had been presented on behalf of Payton. The jury returned with a verdict of death.

On direct appeal, Payton argued that his Eighth Amendment rights were violated because the jury was misled regarding the relevance of his penalty phase evidence. The California Supreme Court, in a 5-2 decision, was unpersuaded. In addressing the constitutional claim, it relied on Boyde v. California, 494 U.S. 370 (1990), where the Supreme Court rejected an argument that the language of factor (k) would not have alerted jurors that they could consider evidence of the defendant's background and character, in addition to information directly linked to the crime. In the context of the entire proceeding, the California Supreme Court found no reasonable likelihood that Payton's jury felt it was precluded from considering his mitigating evidence. Reviewing the claim de novo, a federal district court disagreed with the California Supreme Court's conclusion and granted habeas relief to Payton. A divided three-judge panel in the Ninth Circuit reversed. Rehearing the case en banc, the Ninth Circuit affirmed the district court's grant of habeas relief by a 6-5 vote. The Supreme Court then granted certiorari and remanded for reconsideration in light of Woodford v. Garceau, 538 U.S. 202 (2003), which established that the new limitations on habeas relief contained in amended 28 U.S.C. § 2254(d) applied to Payton's case. That meant habeas relief was barred unless the state court's decision was not merely incorrect, but that is was "contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States." Again, by a vote of 6-5, the Ninth Circuit affirmed the district court. It found that the California Supreme Court had unreasonably applied Supreme Court precedent in holding that factor (k) was not unconstitutionally ambiguous in Payton's case. The Ninth Circuit distinguished Boyde on the ground that it focused on whether a jury would understand that pre-crime evidence about a defendant's background and character fell under factor (k). This was far different, in the appeals court's view, than post-crime evidence such as Payton presented, which it doubted a jury would similarly understand fell within the scope of factor (k). Further, in the Boyde case, the prosecutor did not misstate the law as did the prosecutor in Payton's case, and without correction from the trial court.

Justice Kennedy began the discussion by noting that the Boyde decision had indeed been the correct starting point for the California Supreme Court's analysis. Boyde "established that [factor (k)] does not limit the jury's consideration of extenuating circumstances solely to the circumstances of the crime." In Boyde, the Supreme Court reasoned that factor (k) went to circumstances that might excuse a crime, which would certainly encompass background and character evidence. The California Supreme Court read Boyde as establishing that the text of factor (k) was broad enough to also encompass the post-crime evidence that Payton presented. The failure of the California Supreme Court to distinguish between pre-crime and post-crime evidence was not, in the majority's view, unreasonable. Given that Boyde "held that factor (k) directed consideration of any circumstance that might excuse the crime," it was not "unreasonable to believe that a post-crime character transformation could do so." Indeed, remorse, which is necessarily post-crime, is "something commonly thought to lessen or excuse a defendant's culpability."

Having found that the California Supreme Court reasonably interpreted factor (k) as including post-crime evidence, the majority turned to the question of whether it was nevertheless unreasonable for the state court to conclude that the prosecutor's argument did not mislead the jury about the proper reach of the factor. Even assuming that the California Supreme Court's determination on this point was incorrect, the majority could not deem it unreasonable. In Boyde, the Supreme Court emphasized that an ambiguous instruction must be considered in the context of the entire trial. Here, although the prosecutor affirmatively argued to the jury that it was prohibited from considering Payton's evidence, this came after the jury heard from eight defense witnesses over a two day period without any relevance objection by the prosecutor. The California Supreme Court recognized that, like the Boyde case, for the jury to have accepted the prosecutor's argument would mean it had to believe that the penalty phase served virtually no purpose whatsoever. Further, defense counsel expressly noted the prosecutor's failure to object to the defense evidence in arguing that Payton's post-crime religious conversion was to be considered in deciding what the proper punishment should be. And in addition to asserting that the defense case was irrelevant, the prosecutor also disputed the sincerity of the defense and devoted considerable attention to discounting its importance when compared to the evidence in aggravation. In this discussion, the prosecutor characterized the defense testimony as "evidence" on at least four separate occasions. On this record, "it was not unreasonable for the state court to conclude that the jury believed Payton's evidence was neither credible nor sufficient to outweigh the aggravating factors, not that it was not evidence at all." Additional reasons given for the majority's determination were: (1) the prosecutor's argument was not made in bad faith; (2) the incorrect interpretation was provided to the jury after it had heard the evidence; (3) defense counsel objected to the narrow view of factor (k); and (4) the trial court admonished the jury in response to the objection that comments by the attorneys were not evidence. The majority did concede that the trial court should have confirmed for the jury that Payton's evidence fell within factor (k), but pointed out that the jury was not left without any judicial direction. Of note was an instruction to consider all evidence received during any part of the trial. And the language of factor (k) "directs jurors to consider any other circumstance that might lessen a defendant's culpability." The judgment of the Ninth Circuit was reversed.

Justice Scalia, joined by Justice Thomas, wrote a concurring opinion, reiterating his belief "that limiting a jury's discretion to consider all mitigating evidence does not violate the Eighth Amendment."

Justice Breyer also authored a concurring opinion. Although he found it likely, if he were a California state judge, that he would conclude an Eighth Amendment violation occurred in Payton's case, he was unable to label the California Supreme Court's contrary decision as "unreasonable." Thus, "this is a case in which Congress' instruction to defer to the reasonable conclusions of state-court judges makes a critical difference."

Justice Souter, joined by Justices Stevens and Ginsburg, dissented. Looking to the language in factor (k), Souter concluded that evidence going to the "gravity of the crime" would be commonly understood "as the joint product of intent, act and consequence: intentionally shooting a police officer through the heart is worse than knocking down a pedestrian by careless skateboarding." As for Boyde, he believed it was "coherent with this understanding" of factor (k) to say "that evaluating a defendant's state of mind at the time of the offense can include consideration of his general character and the experiences that affected its development . . .." This is because society, when sitting in judgment, "does not ignore the early hardships of those who turn out bad." But as for a defendant's subsequent experience, Souter found it harder to believe that it would be seen as impacting the seriousness of a crime. That a juror would not have understood that factor (k) included post-crime behavior is shown by the prosecutor's misinterpretation of the statutory language in Payton's case, as well as by the fact that the California Supreme Court in 1983 directed trial courts to include supplementary language for factor (k) in order to ensure that juries were apprised they could consider any other aspect of the defendant's character or record that was proffered as a basis for a sentence less than death. See People v. Easley, 34 Cal.3d 858, 878, n. 10 (Cal. 1983). Later, the standard sentencing instructions were modified to provide that factor (k) included any sympathetic or other aspect of the defendant's character or record that was offered as a basis for a sentence less than death "whether or not related to the offense for which he is on trial." In Souter's view, without such an amendment, "any claim that factor (k) called for consideration of a defendant's personal development in the wake of his crime was simply at odds with common attitudes and the English language."

Souter further pointed to important distinctions between the cases of Payton and Boyde. One difference was the brevity of the defense presentation on behalf of Payton. The testimony took up only 50 pages of transcript compared to the 400 pages in Boyde's case. Second, the prosecutor in Boyde never suggested that the extensive defense evidence concerning Boyde's background and character was beyond the jury's reach. Here, immediately after defense counsel's unsuccessful objection to the prosecutor's mischaracterization of factor (k), the prosecutor "took up exactly where he had left off, arguing that Payton's proffered mitigating evidence could not be considered in the jury's deliberations." Even when the prosecutor was questioning the sincerity of the evidence and arguing that it was outweighed by the aggravation, he continued to erroneously remind the jurors that factor (k) was limited to facts as of the time of the crime. Souter was unable to accept that the prosecutor's error was in any way cured by the trial court's admonishment that the prosecutor's argument was not evidence. "This instruction cured nothing" since the prosecutor's "objectionable comment was not a statement about evidence but a statement of law. Telling the jury that a statement of law was not evidence did nothing to correct its functional error in misstating the law." In addition, there were key differences between the instructions provided to the jury in Payton's and Boyde's cases. In Boyde, the jury was told it "shall" consider all the evidence received during any part of the trial. In Payton's case, the jury was told to consider all the evidence presented during the trial "except as you may be hereafter instructed." The "hereafter," included factor (k).

Souter next observed that even if Payton's penalty trial was "best view the way the majority suggests," that still failed to satisfy Boyde's test of a reasonable likelihood that the jury construed an instruction as foreclosing consideration of mitigation. Souter accused the majority of broadening the holding of Boyde, which was necessary in order to find a reasonable application of the case by the California Supreme Court. By doing so, Souter contended the majority "further obscures the necessarily inexact distinction between cases that are merely wrong and cases with objectively unreasonable error."

Johnson v. United States
544 U.S. ___, 125 S.Ct. 1571 (2005)

Under 28 U.S.C. section §2255, ¶6(4), the one-year limitation period for federal prisoners to file a collateral attack on their sentence begins to run on "the date on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence." In this §2255 case, the Supreme Court addressed the question of how this limitation period applies in a case where the prisoner is attacking his federal sentence on the ground that a state conviction used to enhance that sentence has since been vacated.

After observing that "AEDPA's drafters probably never thought about the situation we face here," the Court (Souter, J., joined by Rehnquist, C.J., and O'Connor, Thomas and Breyer, JJ.), held that a state court's order vacating a federal prisoner's prior conviction falls within the definition of a "fact" within the meaning of §2255 ¶6(4), and that "receiving notice of success [in a collateral challenge to a state court conviction used to enhance a federal sentence] can surely qualify as a kind of discovery falling within the statutory language."

Concerned about the "core purposes" of the AEDPA and the statute's reference to "due diligence," the Court added the additional requirement that, to preserve the ability to rely upon §2255 ¶6(4) in the event the prisoner secures a state court order vacating a conviction used for enhancement, the prisoner must act diligently in initiating the state court proceedings necessary to produce such an order. Characterizing its reasoning as "choosing the bowl of porridge between the one too hot and the one too cold," the Court concluded that a prisoner will be deemed "diligent" in this regard if he begins his attack on the state court conviction(s) used to enhance his federal sentence at the time the federal judgment against him is entered. In footnote 8, the Court noted the following qualifications and exceptions:

Applying these rules to Johnson, the Court affirmed the Eleventh Circuit's determination that his §2255 challenge to an eight year enhancement to his federal sentence based on a subsequently vacated Georgia conviction was untimely because Johnson had waited more than three years after entry of judgment in his federal case to begin his successful collateral attack on the state conviction.

Justice Kennedy (joined by Stevens, Scalia and Ginsburg, JJ.) dissented, agreeing with the Court's determination that a state court vacatur order does constitute a "fact" capable of discovery within the meaning of §2255 ¶6(4), but pointedly disagreeing with the Court's addition of the "diligence" requirement. Justice Kennedy summarized his disagreement as follows:

As to the last point in his summary, Justice Kennedy went on to note that "[t]he law, and the decisions of this Court, put extraordinary demands on defense counsel," and that "[i]t is most troubling for a Court that insists on high standards of performance for defense counsel now to instruct that collateral proceedings must be commenced in one or more States during the critical time immediately after judgment and before appeal."

Rhines v. Weber
544 U.S. ___, 125 S.Ct. 1528 (2005)

Addressing “the problem of a ‘mixed’ petition for habeas corpus relief in which a state prisoner presents a federal court with a single petition containing some claims that have been exhausted in the state courts and some that have not,” the Supreme Court held that the Eighth Circuit “erred to the extent it concluded that stay and abeyance is always impermissible,” and remanded petitioner’s case for further proceedings. Writing for the Court, Justice O’Connor (joined by Rehnquist, C.J., and Stevens, Scalia, Kennedy, Thomas, Ginsburg, and Breyer, JJ), began by outlining the difficulties brought about by the combination of the “total exhaustion” rule of Rose v. Lundy, 455 U.S. 509 (1982), the §2244(d) statute of limitations enacted in 1996, and the Court’s holding in Duncan v. Walker, 533 U.S. 167 (2001), that the limitations period is not tolled by the pendency of a federal habeas petition. “If a petitioner files a timely but mixed petition in federal district court,” the Court explained, “and the district court dismisses it under Lundy after the limitations period has expired, this will likely mean the termination of any federal review.”

“[R]ecogniz[ing] the gravity of this problem and the difficulty it has posed for petitioners and federal district courts alike,” the Court endorsed the use of stay and abeyance, but made clear that it “should be available only in limited circumstances.” The Court then set forth the following guidelines as conditions:

(internal citations omitted). The Court concluded by vacating the Eighth Circuit’s judgment and remanding for a determination of “whether the District Court’s grant of a stay in this case constituted an abuse of discretion.”

Justice Stevens (joined by Ginsburg and Breyer, JJ.) concurred “on the understanding that [the Court’s] reference to ‘good cause’ for failing to exhaust state remedies more promptly is not intended to impose the sort of strict and inflexible requirement that would ‘trap the unwary pro se prisoner.’” (quoting Rose v. Lundy).

Justice Souter (joined by Ginsburg and Breyer, JJ.), concurred in part and in the judgment, but wrote separately to express his view that, “[i]nstead of conditioning stay-and-abeyance on ‘good cause’ for delay, I would simply hold the order unavailable on a demonstration of ‘intentionally dilatory litigation tactics.’” “I fear,” Justice Souter went on to explain, “that threshold enquiries into good cause will give the district courts too much trouble to be worth the time; far better to wait for the alarm to sound when there is some indication that a petitioner is gaming the system.”

Bell v. Cone
543 U.S. ___, 125 S.Ct. 847 (2004) (per curiam)

In a per curiam opinion, the Supreme Court reversed the Sixth Circuit in a Tennessee death penalty case, ruling that 28 U.S.C. § 2254(d) precluded relief on Cone’s claim that the “especially heinous, atrocious, or cruel” (HAC) aggravating factor was unconstitutionally vague, and that the Tennessee Supreme Court failed to cure the constitutional deficiency on appeal.

Cone killed an elderly couple in their home at the conclusion of a two-day crime spree. Both victims were “repeatedly beaten about the head until they died.” State v. Cone, 665 S.W.2d 87, 90-91 (Tenn. 1984). When the bodies were discovered, they were “horribly mutilated and cruelly beaten.” Id. at 90. The jury found four aggravating circumstances: (1) conviction of one or more violent felonies; (2) the murders were “especially heinous, atrocious, or cruel in that they involved torture or depravity of mind”; (3) the murders were committed for the purpose of preventing a lawful arrest or prosecution; and (4) the knowing creation of a risk of death to two or more persons, other than the victim murdered, during the murder. Finding that the aggravating circumstances outweighed those in mitigation, the jury sentenced Cone to death.

On direct appeal, the Tennessee Supreme Court struck the fourth aggravating circumstance, finding that the danger Cone had posed to others was not sufficiently close in time to the murders. As for the other three aggravators, the state supreme court found they were clearly shown by the evidence. Regarding the HAC aggravating circumstance, the court recounted evidence that the victims had been:

Id. at 94-95. Cone’s conviction and death sentence were affirmed.

After further state court litigation, Cone filed a second amended petition for post-conviction relief which included a claim that the HAC aggravating circumstance was unconstitutionally vague. The trial court ruled that the claim had been previously rejected by the state supreme court on direct appeal or by the intermediate appellate court when it ruled on Cone’s first post-conviction relief petition. The Tennessee Court of Criminal Appeals affirmed the denial of relief and the Tennessee Supreme Court denied Cone permission to appeal.

In federal habeas proceedings, the Sixth Circuit Court of Appeals rejected the State’s contention that Cone’s HAC challenge was procedurally defaulted. The court of appeals concluded that a Tennessee statute requiring review of every death sentence by the Tennessee Supreme Court necessarily included the consideration of constitutional deficiencies in the aggravating circumstances. Therefore, the court reasoned, Cone’s claim had been fairly presented to the highest state court even if it had not been expressly raised by Cone himself during the direct appeal. The Warden argued that this conclusion was in tension with other circuit courts which have held that a habeas petitioner must expressly present a constitutional challenge in order to achieve exhaustion even if mandatory review is part of state law. The Supreme Court found it unnecessary to express a view on this issue, noting that 28 U.S.C. § 2254(b)(2) allows a federal court to deny a claim on the merits notwithstanding its lack of exhaustion. The Court did emphasize, however, that

Bell v. Cone, at 850 n. 3.

As for the merits of the claim, the Sixth Circuit ruled that the state court’s affirmance of Cone’s death sentence in light of the HAC aggravating circumstance was “contrary to” Godfrey v. Georgia, 446 U.S. 420 (1980). Although the HAC instruction in Cone’s case included language that had not been directly addressed in any of the Supreme Court cases concerning vagueness challenges to aggravators, the Sixth Circuit nevertheless found the HAC circumstance unconstitutional in light of Godfrey, Maynard v. Cartwright, 486 U.S. 356 (1988), Walton v. Arizona, 497 U.S. 639 (1990), and Shell v. Mississippi, 498 U.S. 1 (1990) (per curiam). While the latter three cases post-dated Cone’s direct appeal, the Sixth Circuit determined that they did not represent new law, but were dictated by Godfrey. In its opinion, the Supreme Court assumed, without deciding, that the court of appeals was correct about this point.

Finally, the Sixth Circuit concluded that the state supreme court failed to cure the vagueness when it reviewed the aggravator on direct appeal by using the narrowing construction it previously adopted in State v. Dicks, 615 S.W.2d 126 (1981). The Sixth Circuit noted that neither the language in Dicks was included in the state court opinion, nor was the case cited. Rather, in the view of the Sixth Circuit, the state court simply found that the labels “heinous, atrocious, or cruel,” without more, applied to the facts of this case.

In addressing whether the state court had in fact cured any vagueness problem the Supreme Court acknowledged in a footnote that Ring v. Arizona, 536 U.S. 584 (2002) now requires that a jury determine the presence of aggravating circumstances. Whether a state appellate court is still permitted to cure an aggravator by applying a narrower construction than provided to the jury was not addressed in this case because Ring is not retroactive.

As for the Sixth Circuit’s conclusion that the state court failed to apply a narrowing construction, the Supreme Court began by stating:

Bell v. Cone, at 853. It then rejected as baseless the Sixth Circuit’s conclusion that the state court “simply, but explicitly, satisfied itself that the labels 'heinous, atrocious, or cruel,' without more, applied” to the murder. The Supreme Court pointed out that the Tennessee Supreme Court had narrowly construed the HAC aggravator and followed that precedent numerous times prior to its affirmance of Cone’s sentence. “[A]bsent an affirmative indication to the contrary,” the Court had to “presume that [the state supreme court] did the same thing here.” Id.

Even without any presumption, the Supreme Court would find that a narrowing construction had been applied by the Tennessee Supreme Court in Cone’s case. This is because the facts the state court relied upon to uphold the HAC aggravator in Cone’s case closely tracked the facts found in other cases to satisfy the torture and depravity prongs of the narrowed HAC aggravator. Additional support for the Supreme Court’s conclusion is found in the direct appeal briefing where both Cone and the State relied on cases applying a narrowed construction of the HAC aggravator.

Finally, the Supreme Court ruled that the narrowed constructed of the HAC that had been adopted by the Tennessee Supreme Court – a conscienceless or pitiless crime which is unnecessarily torturous to the victim – was not contrary to Supreme Court precedent. Indeed, this same language had been approved by the Supreme Court in Proffitt v. Florida, 428 U.S. 242 (1976). That Tennessee’s HAC aggravating circumstance, as construed by the state supreme court, could meaningfully narrow the class of death eligible defendants was demonstrated by the state supreme court’s application of HAC in later cases where it was explained that the HAC required either evidence that the defendant inflicted torture on the victim before death or that the defendant committed acts evincing a depraved state of mind. The state court also clarified that torture could not be established when the death was instantaneous and depravity of mind was not shown simply because the victim was shot twice.

Because the state court decision affirming Cone’s sentence was not “contrary to” Supreme Court precedent, the Court held that the Sixth Circuit was powerless to grant habeas relief.

Justice Ginsburg concurred, joined by Justices Souter and Breyer. She “agree[d] with the Court that, once the highest court of a State has dispositively decided a point of law, it is not incumbent on that court to cite its precedential decision in every case thereafter presenting the same issue in order to demonstrate its adherence to the pathmarking decision.” She wrote separately to explain her view that the Court’s decision in this case was confined to a situation where a state court has “confronted and decided an issue governed by a prior ruling.” Id. at 856. It did not, she emphasized,

Id. at 856.

Pace v. DiGuglielmo
544 U.S. ___, 125 S.Ct. 1807 (2005)

In a 5-4 decision, the Supreme Court held that a state post-conviction petition that is rejected by the state court as untimely is not “properly filed” for purposes of tolling the limitation period for federal habeas petitions under 28 U.S.C. §2244(d)(2). The majority decision was authored by the Chief Justice, and joined by Justices O’Connor, Scalia, Kennedy and Thomas. Justice Stevens filed a dissenting opinion which was joined by Justices Souter, Ginsburg and Breyer.

In February 1986, John Pace pleaded guilty to several crimes in Pennsylvania and was sentenced to life imprisonment without the possibility of parole. He neither moved to withdraw his guilty plea nor filed a direct appeal. In August of 1986, Pace did seek post-conviction relief in state court. Those proceedings were completed in September 1992.

In April 1996, through the Antiterrorism and Effective Death Penalty Act of 1996 (AEPDA), Congress created a 1-year statute of limitations for state prisoners filing federal habeas corpus petitions. 28 U.S.C. §2244(d)(1). The limitation period is tolled, however, while “a properly filed application for State post-conviction or other collateral review with respect to the pertinent judgment or claim is pending.” 28 U.S.C. §2244(d)(2).

On November 27, 1996, Pace filed another state postconviction application under a relatively new law, the Pennsylvania Post Conviction Relief Act (PCRA). The PCRA had been amended in 1995 to include a statute of limitations running from finality of the conviction. Although the statute contained three exceptions, Pace said nothing in his application about why his petition was timely despite being filed more than one year after his conviction became final. In briefing, Pace later argued that the time limit did not apply to him. The state court ultimately dismissed the petition as untimely, noting that Pace neither alleged nor proved that he fell within one of the three exceptions to the one-year limitation period. State proceedings were completed on July 29, 1999.

Pace filed his federal habeas petition on December 24, 1999. The district court found that the petition was timely. Although the petition was “due” in April 1997, one year after enactment of the AEDPA limitation period, the district court found that Pace was entitled to tolling under §2244(d)(2), for the time his PCRA application was pending. Looking to the exceptions to the limitation period found in the PCRA, the district concluded that the time limit was not a “condition to filing,” but rather was a “condition to obtaining relief” as described in Artuz v. Bennett, 531 U.S. 4, 11 (2000). As such, the PCRA application was “properly filed” even though Pace was not entitled to relief in state court because of his untimeliness. In the alternative, the district court found that Pace was entitled to equitable tolling.

The Third Circuit Court of Appeals reversed. It found that the PCRA time limit was a “condition to filing” and, therefore, Pace’s untimely application was not “properly filed.” The appeals court further found that there were no extraordinary circumstances justifying equitable tolling. Certiorari was granted.

The Chief Justice began the Court’s analysis by discussing Artuz v. Bennett. There, the Court held that time limits on postconviction petitions are conditions to filing. 531 U.S. at 8, 11. The Court reserved, however, the question of whether the existence of exceptions to a time limit could prevent a late application from being considered improperly filed. Now addressing that issue, the Court could “see no grounds for treating the two differently.” Pace, 2005 WL 957194, *4. Applying a common understanding of “properly filed,” there was no obvious distinction between an untimely petition that failed to meet any exceptions to the state time limit and an untimely petition under a system that allowed for no exceptions. Further, the Court concluded the purpose of AEDPA’s limitation period confirmed the Court’s “commonsense reading” of the statute. Accepting Pace’s interpretation of §2244(d)(2), a state prisoner could receive a de facto extension of time simply by filing an untimely state postconviction application which is contrary to what AEDPA sought to achieve and also invites abusive delay.

Pointing to this same conclusion about the correct meaning of “properly filed application” was the Court’s decision in Carey v. Saffold, 536 U.S. 214 (2002). There, in the context of addressing when a postconviction petition is “pending,” the Court observed that if the state supreme court had clearly ruled that the prisoner’s habeas petition was untimely, that would have been “the end of the matter,” regardless of whether the state court also considered the merits of the claim or whether its timeliness ruling was intertwined with the merits. What was intimated in Saffold was expressly held here: “When a postconviction petition is untimely under state law, ‘that [is] the end of the matter’ for purposes of §2244(d)(2).” Pace, 2005 WL 957194, *4, quoting Saffold, 536 U.S. at 226.

The Court was unpersuaded by Pace’s contention that “conditions to filing” are limited to those conditions necessary to get a clerk to accept the application, in contrast to conditions which require some judicial consideration. The Court pointed out that some of the “conditions to filing” delineated in Artuz, such as jurisdictional matters, often require judicial scrutiny. Looking to Pennsylvania law, the Court further observed that there were two mandatory conditions for filing a postconviction application: (A) the petition “shall” be filed within the time limit, and (B) the proceedings “shall be initiated by filing” a verified petition and “3 copies with the clerk of the court in which the defendant was convicted and sentenced.” The natural reading of this rule was that “(A) is every bit as much of a ‘condition to filing’ as (B).” Id.

Pace had also asserted that because §2244(d)(2) refers to a “properly filed application,” that any condition that must be applied on a claim-by-claim basis cannot constitute a “condition to filing.” The Court found this contention refuted by §2244 itself. In section 2244(b)(3)(C), for example, a court of appeals is permitted to authorize the filing of a second or successive “application” but only if the appeals court determines that the “application” makes a prima facie showing that the “application” satisfies the requirements of the subsection. The requirements of the subsection, however, necessitate inquiry into specific claims, not a review of the application as a whole. In addition, §2244(d)(1) states that the 1-year limitation period applies to an “application” for a writ of habeas corpus. That subsection goes on to provide the means for calculating the limitation with respect to the application as a whole – (A) (date of final judgment) – but then the three other potential starting dates contained in (B), (C) and (D) require claim-by-claim consideration. Indeed, Pace’s argument was inconsistent with §2244(d)(2) itself which refers to a “properly filed application . . . with respect to the pertinent judgment or claim.” (Emphasis added.)

Pace’s final argument was that it would be unfair to interpret the statute in the manner adopted by the Court because state prisoners may not know in advance whether a postconviction application will be timely or not. A prisoner could spend years litigating in state court in a good faith effort to exhaust his or her remedies only to find at the end of the process that the application was never “properly filed,” and thus a federal habeas petition would be time batted. The Court’s response to this was noting that a prisoner who faced uncertainty about whether a state postconviction application would be found untimely could file a “protective” petition in federal court and ask the federal court to stay the federal proceedings until state court remedies were exhausted. See Rhines v. Weber, 125 S.Ct. 1528. The Court explained, “[a] petitioner’s reasonable confusion about whether a state filing would be timely will ordinarily constitute ‘good cause’ for him to file in federal court.” Pace, 2005 WL 957194, *5.

The majority rejected the dissent’s contention that Artuz’s ruling that the state procedural bars at issue there did not constitute “conditions to filing” required a conclusion that the PCRA time limit was also not a filing to condition. The dissent, according to the majority, ignored the fact that in Artuz the Court distinguished between time limits and procedural bars. Noting that Pennsylvania law describes the time limit as “jurisdictional,” and that procedural bars similar to those at issue in Artuz are separately delineated in a section on “eligibility for relief,” the majority found: “it must be the case that a petition that cannot even be initiated or considered due to the failure to include a timely claim is not ‘properly filed.’” Id. In conclusion, the Court expressly held “that time limits, no matter their form, are ‘filing’ conditions.”

Ruling that Pace was not entitled to statutory tolling, the Court turned to the question of equitable tolling. Because DiGuglielmo assumed that equitable tolling applied to AEDPA’s statute of limitations, and because the Court was convinced Pace was not entitled to such tolling under any standard, it “assume[d] without deciding its application for purposes of this case.” ” Pace, 2005 WL 957194, *6 n.8. In general, to receive equitable tolling a litigant must establish: (1) he diligently pursued his rights; and (2) some extraordinary circumstance stood in his way. As for the second element, Pace contended he was trapped by Third Circuit precedent which required him to exhaust state remedies where there was a chance of a merits review, and state law which made it appear he could gain relief despite a state application’s untimeliness. The Court did not decide whether this situation constituted an “extraordinary circumstance,” relying instead on Pace’s lack of diligence to reject his request for equitable tolling. Two of the claims raised by Pace in his PCRA petition were available to him as early as 1986. The third claim related to events occurring in or before 1991. Despite the availability of the claims, Pace “waited years, without any valid justification to assert these claims in his November 27, 1996, PCRA petition.” In addition to sitting on his rights before filing the PCRA petition, the Court noted that Pace also sat on them for five more months after the PCRA proceedings ended before filing his federal habeas petition. “Under long-established principles, [Pace’s] lack of diligence precludes equity’s operation.” Pace, 2005 WL 957194, *6.

In dissent, Justice Stevens contended that the interpretation of §2244(d)(2) adopted by the majority was “not compelled by the text of that provision and will most assuredly frustrate its purpose.” Pace, 2005 WL 957194, *7. He pointed out that in Artuz, the Court held that a state postconviction application was properly filed even if the legal claims contained in the application were procedurally barred under state law. The question here was whether the untimeliness finding was more akin to a procedural bar than a general precondition to filing. Before addressing that, Stevens discussed why Pace’s PCRA petition was held to be untimely.

Prior to initiating federal habeas proceedings, which Pace had to do by April 24, 1997, Pace was required to exhaust any available state remedies. Pace filed his PCRA application in November of 1996. In order to receive merits review by the state court, Pace had to demonstrate that his claims were not procedurally barred and so he cited to state cases providing judicial exceptions to procedural default. At the time Pace filed his PCRA petition, no Pennsylvania court had applied the new statute of limitations in a case where the petitioner’s conviction had become final prior to the effective date of the Act. Pace was denied relief on the merits by the lower court after the AEDPA limitation period had expired. When Pace appealed, the Commonwealth argued for the first time that the application was untimely under the PCRA statute of limitations. The state appellate court agreed, looking only to the statutory exceptions to untimeliness contained in the PCRA.

After providing this background, Stevens turned back to Artuz. In that decision, the Court refused to interpret “properly filed application” to mean “application raising claims that are not mandatorily procedurally barred.” 531 U.S. at 9. The Court reasoned that to do so would elide “the difference between an ‘application’ and a ‘claim.’ Only individual claims and not the application containing those claims, can be procedurally defaulted under state law . . ..” Id. The Court went on in Artuz to condemn the “verbal gymnastics” that would be required under the rejected interpretation of “properly filed application,” noting that a judge presumably would have to find an application “properly filed” as to nonbarred claims, but improperly filed as to claims that were defaulted. Nothing in the statute “contain[ed] the peculiar suggestion that a single application can be both ‘properly filed’ and not ‘properly filed.’” Id., at 10.

In Stevens’ view, the same reasoning applied here with equal force. He posited a situation where one claim in a PCRA application received a merits denial while another claim was denied as untimely. Under the majority’s rule, the first claim would be considered a “properly filed application,” and the second would not. Stevens’ concluded that “[t]his is precisely the type of incoherent result that Artuz sought to avoid.” Pace, 2005 WL 957194, *9.

Stevens went on to criticize the majority for suggesting that the phrase “properly filed” means something different when applied to time bars than it does in the context of other procedural bars. This was inconsistent with the Court’s general refusal “to adopt rules that would give the same statutory provision different meanings in different contexts.” Pace, 2005 WL 957194, *10.

Stevens would simply apply Artuz’s rule to time bars that operate like a procedural default, which he found to be consonant with the statutory text. This approach, he concluded, would accomplish the purposes behind AEDPA As the Court found in Duncan v. Walker, 533 U.S. 167, 180 (2001), §2244(d)(2) was designed to provide a strong incentive for state prisoners to seek relief in state court before initiating federal habeas proceedings. The majority’s rule, according to Stevens, is unfaithful to the goal of streamlining and simplifying the habeas system and avoiding overburdening federal courts by encouraging piecemeal litigation.

The majority justified its rule by noting that a different interpretation would allow indefinite extension of the federal statute of limitations through the filing of untimely state postconviction applications. Stevens argued that the fear of such a practice by prisoners was misguided for two reasons. First, a prisoner’s principle interest is in obtaining speedy federal review of his claims. Second, “properly filed” could be defined “as excluding any filings deemed by the district court to be repetitious and abusive.” Pace, 2005 WL 957194, *11. This would prevent state prisoners from prolonging indefinitely the federal state of limitations.

Stevens ended by predicting that the majority’s rule will most likely increase delays in the federal system in the wake of a “flood of protective filings in the federal district courts.” Id.

Roper v. Simmons
543 U.S. ___, 125 S.Ct. 1183 (2005)

In a 5-4 decision, the Supreme Court ruled that the Eighth Amendment ban on cruel and unusual punishment prohibits executing persons who were under the age of 18 at the time they committed a capital offense. The majority opinion was written by Justice Kennedy, and joined by Justices Stevens, Souter, Ginsburg and Breyer. Justice Stevens wrote a concurrence, which was joined by Justice Ginsburg. Justice O’Connor authored a dissent. Justice Scalia also dissented, joined by the Chief Justice and Justice Thomas.

In Stanford v. Kentucky, 492 U.S. 361 (1989), a divided Supreme Court ruled that the Constitution did not bar execution of individuals who were 16 or 17 at the time they committed a capital crime. In this Missouri death penalty case, the Court revisited that decision.

The respondent in this case, Christopher Simmons, was 17 and a junior in high school when he told friends that he wanted to murder someone. Simmons proposed to two friends that they break into a house, tie up the victim, and throw the person off a bridge. Simmons assured his friends that they could get away with the crime because they were minors. One of the friends backed out of the plan and left before Simmons and the other boy entered the home of Shirley Crook. When Mrs. Crook encountered the two young men, Simmons recognized her from a prior automobile accident. He later admitted that this coincidence confirmed his resolve to kill her. The victim was bound and taken to a state park. Her entire face wrapped in duct tape, Mrs. Crook was thrown from a bridge into the Meramec River where she drowned. Witnesses testified that Simmons bragged about the killing later.

Simmons was arrested the following day. Not only did he confess, he agreed to perform a videotaped reenactment at the crime scene. He was tried as an adult pursuant to state law and convicted of capital murder. Although his defense attorney argued that his young age should be considered as a factor in mitigation, the prosecutor argued to the contrary. The jury recommended a sentence of death which the trial court imposed. The Missouri Supreme Court affirmed Simmons’ convictions and sentence, as well as the denial of an initial motion for postconviction relief.

Following the decision in Atkins v. Virginia, 536 U.S. 924 (2002), where the Supreme Court ruled that it would be unconstitutional to execute a mentally retarded inmate, Simmons again sought post-conviction relief, arguing that the reasoning of Atkins established it would be unconstitutional to execute someone who was under 18 at the time of the murder. The Missouri Supreme Court agreed, State ex rel. Simmons v. Roper, 112 S.W.3d 397 (2003), and the Supreme Court granted Missouri’s certiorari petition.

Justice Kennedy began by looking to Thompson v. Oklahoma, 487 U.S. 815 (1988), where a plurality of the Court concluded that the evolving standards of decency did not permit execution of an inmate who was under the age of 16 at the time of the crime. In arriving at that conclusion, the plurality looked to state statutes, as well as the views expressed by respected professional organizations, foreign nations sharing the Anglo-American heritage, and leading members of the Western European community. The plurality also noted that it was increasingly rare for juries to impose death sentences on those who were younger than 16 at the time of their crime. Finally, the plurality brought its independent judgment to bear on the permissibility of executing someone who was only 15 when he committed the murder. It concluded that such a person was both less morally culpable and less likely to be deterred by the existence of the death penalty. Justice O’Connor concurred, although on narrower grounds.

The next year, the Court decided Stanford v. Kentucky, 492 U.S. 361 (1989). In a 5-4 decision, the Court concluded that the prevailing standards of decency in the United States did not preclude imposition of the death penalty on 16 and 17-year-olds who committed capital murder. The Court observed that 22 of the 37 death penalty jurisdictions permitted the death penalty for 16-year-old offenders, and 25 of those jurisdictions allowed it for 17-year-old offenders. This failed to demonstrate a national consensus against executing all juvenile offenders. A plurality of the Court emphatically rejected the suggestion that the Court should bring its own judgment to bear on the constitutional question. That very same day, the Court ruled in Penry v. Lynaugh, 492 U.S. 302 (1989), that the Eighth Amendment did not bar the execution of the mentally retarded. In reaching that conclusion, the Court stressed that only two death penalty jurisdictions banned execution for that class of defendants.

In 2002, the Court revisited the question of whether it was cruel and unusual to execute mentally retarded individuals. In finding that the standards of decency had evolved such that it would violate the Eighth Amendment to execute the mentally retarded, the Court noted “objective indicia of society’s standards, as expressed in legislative enactments and state practice with respect to executions of the mentally retarded.” Simmons, at 1191. Observing that the practice had become “truly unusual,” the Court in Atkins discerned a national consensus against it. The analysis did not stop there, however. Notably, the Atkins Court returned to the pre-Stanford rule that the Court would ultimately bring to bear its own judgment on the question of the acceptability of the death penalty under the Eighth Amendment. The Court then considered the diminished culpability harbored by the mentally retarded, a belief that it is less defensible to impose the death penalty as retribution for a crime committed by a mentally retarded individual, and that the death penalty is less likely to have a deterrent effect on the mentally retarded. These factors, combined with the national consensus against executing the mentally retarded, led the Court to find an Eighth Amendment prohibition against the practice.

Using the Atkins approach as the model, the Court turned to the question of whether there was now a national consensus against executing juveniles. It identified 30 States that prohibit juvenile executions, the same number that prohibited the execution of the mentally retarded at the time Atkins was decided. Further paralleling Atkins was the infrequency of executions of the juveniles by States where the practice was permitted. Post-Penry, only five States executed mentally retarded individuals. Post-Stanford, there were six States that executed inmates for crimes committed as juveniles. In the last ten years, there were only three. Also noted by the Court was that Kevin Stanford’s death sentence had been commuted by the Governor of Kentucky based on Stanford’s age at the time of the crime.

Differing from Atkins was the rate of abolition. Sixteen States that had permitted execution of the mentally retarded at the time of Penry had abandoned that practice by the time the Supreme Court revisited Atkins’ Eighth Amendment challenge. In contrast, only five States had altered the rule about executing juveniles following Stanford. Despite this less dramatic numerical change, the Court still found it significant given the consistency of the direction of change and the absence of any State reinstating a juvenile death penalty after Stanford. The trend towards abolition of the juvenile death penalty was especially notable given the prevalence of anticrime legislation in the past decades, much of which sought to crack down on juvenile crime.

The Court also observed that the slower pace of abolition concerning the juvenile death penalty could be explained by the fact that more death penalty States had already prohibited the practice at the time of Stanford than had barred executing the mentally retarded at the time of Penry. “If anything, this shows that the impropriety of executing juveniles between 16 and 18 years of age gained wide recognition earlier than the impropriety of executing the mentally retarded.” Simmons, at 1193.

In arguing that there was not a national consensus against executing juveniles, Missouri had pointed to the United States Senate’s action when it ratified the International Covenant on Civil and Political Rights. Although that treaty banned capital punishment for juveniles, the Senate’s ratification was subject to the President’s proposed reservation from that particular provision. The Court was unimpressed with this fact, pointing out that since this reservation occurred, five States abandoned capital punishment for juveniles. Further, Congress later addressed the propriety of executing juveniles when it enacted the Federal Death Penalty Act in 1994 and determined that the death penalty should not extend to them.

The Court then held that a ban on the execution of juveniles is required by the Eighth Amendment. In so doing, it discussed its Eighth Amendment rules which “vindicate the underlying principle that the death penalty is reserved for a narrow category of crimes and offenders.” Id., at 1195. The Court found three general differences between juveniles under 18 and adults which demonstrated that “juvenile offenders cannot with reliability be classified among the worst offenders.” First, juveniles lack maturity and a fully developed sense of responsibility, often resulting in impetuous and ill-considered actions. Notably, almost every State prohibits persons under age 18 from voting, serving on juries, or marrying without parental consent. Second, juveniles are more vulnerable to negative influences and outside pressures. Third, a juvenile’s character is not as well formed as an adult. These key differences “render suspect any conclusion that a juvenile falls among the worst offenders.” Id.

Given the diminished culpability of juveniles, the Court found that the penological justifications for the death penalty applied with lesser force to juveniles than to adults. With diminished blameworthiness, the case for retribution declines. Also, the characteristics of juveniles that make them less culpable suggest that juveniles will be less susceptible to deterrence. And to the extent that punishment does provide a deterrent effect with juveniles, the punishment of life imprisonment without the possibility of parole is certainly a severe sanction, particularly to a young person.

The Court acknowledged that there could be rare instances where juveniles possessed sufficient psychological maturity and demonstrated sufficient depravity to merit a death sentence. But that did not justify rejecting a categorical rule and instead letting juries weigh youth as mitigation. The Court explained:

Simmons, at 1197.

In concluding that Stanford should no longer be deemed controlling on this issue, the Court identified three reasons: (1) objective indicia of consensus has changed since 1989; (2) the Stanford Court should have included non-death penalty jurisdictions in its analysis; and (3) to any extent that Stanford was based on a rejection of the idea that the Court must bring its independent judgment to bear on the proportionality of the death penalty for a class of offenders, that rejection was inconsistent with prior Eighth Amendment decisions.

The Court went on to note that its ruling in this case “finds confirmation in the stark reality that the United States is the only country in the world that continues to given official sanction to the juvenile death penalty.” Simmons, at 1198. While this is not a controlling fact, the Court noted that it has for many, many years referred to international authorities as instructive for its interpretation of the Eighth Amendment.

Justice Stevens concurred to note the majority opinion’s reaffirmation of the evolving standards of decency as the guiding principle in construing the Eighth Amendment. Stevens observed that at the time the Constitution was drafted, the execution of 7 year old children was permissible. If construction of the Constitution was frozen in time, there would be no impediment to such a practice today. Given the evolving standards, Stevens expects that the drafters of the Constitution, were they sitting today, would join Justice Kennedy’s opinion.

Justice O’Connor wrote a lengthy dissent in which she expressed her belief that “[n]either the objective evidence of contemporary societal values, nor the Court’s moral proportionality analysis, nor the two in tandem suffice to justify [the Court’s] ruling.” Simmons, at 1206. O’Connor also criticized the majority “for failing to reprove, or even to acknowledge, the Supreme Court of Missouri’s unabashed refusal to follow [the Court’s] controlling opinion in Stanford.” Simmons, at 1209. This was clear error, according to O’Connor because only the Supreme Court itself can overrule one of its precedents. O’Connor concluded that the affirmance of the state court’s judgment, “without so much as a slap on the hand,” “threatens to invite frequent and disruptive reassessments of [the Court’s] Eighth Amendment precedents.” Id., at 1209-1210.

Although disagreeing with the majority’s finding that the international consensus against juvenile executions confirmed that a categorical ban on the practice was required, O’Connor rejected Justice Scalia’s contention in his dissent that foreign and international law have no place in the Court’s Eighth Amendment jurisprudence. She noted that the Court had for nearly half a century “consistently referred to foreign and international law as relevant to [the Court’s] assessment of the evolving standard of decency.” Id., at 1215.

Justice Scalia also penned a lengthy dissent. In summary, he stated:

The Court reaches [its] implausible result by purporting to advert, not to the original meaning of the Eighth Amendment, but to "the evolving standards of decency," of our national society. It then finds, on the flimsiest of grounds, that a national consensus which could not be perceived in our people's laws barely 15 years ago now solidly exists. Worse still, the Court says in so many words that what our people's laws say about the issue does not, in the last analysis, matter: "[I]n the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment." The Court thus proclaims itself sole arbiter of our Nation's moral standards--and in the course of discharging that awesome responsibility purports to take guidance from the views of foreign courts and legislatures. Because I do not believe that the meaning of our Eighth Amendment, any more than the meaning of other provisions of our Constitution, should be determined by the subjective views of five Members of this Court and like-minded foreigners, I dissent.

Id., at 1217 (citations omitted.) Like Justice O’Connor, Scalia also complained about the majority’s failure to even admonish the Missouri Supreme Court “for its flagrant disregard of [the Court’s] precedent in Stanford.” Id., at 1229.

Deck v. Missouri
544 U.S. ___, 125 S.Ct. 2007 (2005)

In a 7-2 decision, the Supreme Court held that the United States Constitution forbids the use of visible shackles at the guilt and penalty phases of a capital trial, unless that use is justified by “an essential state interest,” such as courtroom security, that is specific to the defendant on trial. Where a defendant has been impermissibly shackled, the prosecution bears the burden of proving that the shackling was harmless beyond a reasonable doubt. Justice Breyer wrote the decision for the Court. Justice Thomas filed a dissenting opinion, which was joined by Justice Scalia.

Carman Deck was tried in Missouri for the killing of an elderly couple during a robbery. At the trial, Deck was required to wear leg braces that apparently were not visible to the jury. He was convicted of the double murder and sentenced to death. On appeal, the state supreme court set aside the death sentence. At the penalty phase retrial, over defense counsel’s objections, Deck was shackled with leg irons, handcuffs and a belly chain. When defense counsel complained that the shackles would make Deck appear to the jurors to remain a violent person, the trial court countered with the view that the shackles took fear out of the jurors’ minds. Deck was again sentenced to death.

On appeal, the Missouri Supreme Court rejected a constitutional challenge to the shackling. It concluded that the trial court properly exercised its discretion to employ the restraints given: (1) there was no record of the extent to which the jury was aware of the shackles; (2) there was no claim that the shackles impeded Deck’s participation in the proceedings; and (3) there was evidence that Deck presented a risk of fleeing given that he was a repeat offender who may have killed the elderly couple to avoid being returned to custody. Further, the state supreme court found that Deck failed to demonstrate prejudice from the shackling.

The Supreme Court began its analysis of Deck’s claim of constitutional error by asking whether the Constitution permits a State to use visible shackles during the guilt-innocence phase of a trial. The Court’s answer was: “The law has long forbidden routine use of visible shackles during the guilt phase; it permits a State to shackle a criminal defendant only in the presence of a special need.” Deck, 2005 WL 1200394, *3. The Court found this rule to be deeply rooted in the common law and followed by American courts. Although there has been disagreement about the degree of discretion afforded to trial courts in determining whether there is a special need for restraints, the American courts “settled virtually without exception on a basic rule embodying notions of fundamental fairness: trial courts may not shackle defendants routinely, but only if there is a particular reason to do so.” 2005 WL 1200394, *4. In more recent times, the Supreme Court itself has “suggested that a version of this rule forms part of the Fifth and Fourteenth Amendments’ due process guarantee.” Id. In Illinois v. Allen, 397 U.S. 337, 343-344 (1970), for example, the Court considered whether the Constitution permits special measures, including physical restraints, with a particularly obstreperous defendant. The Court concluded that binding and gagging could be the fairest and most reasonable solution to controlling such a defendant and the proceedings. In so ruling, the Court added that “even to contemplate such a technique . . . arouses a feeling that no person should be tried while shackled and gagged except as a last resort.” Id., at 344.

Subsequently, in Holbrook v. Flynn, 475 U.S. 560 (1986), the Court considered the constitutionality of a special security arrangement that involved having uniformed security personnel seated in the first row of the courtroom’s spectator section. In finding no constitutional violation, the Court observed that this use of security personnel was not “the sort of inherently prejudicial practice that, like shackling, should be permitted only where justified by an essential state interest specific to each trial.” Id., at 568-569.

Lower courts had taken the statements from these decisions as setting forth a constitutional standard that embodies the common law rule. The ensuing disagreement among the lower courts concerned only the procedural steps necessary for a court to take in determining whether to order restraints, the type of evidence needed to justify shackles, and what sort of prejudice would warrant a new trial. Given this history, the Court concluded that its prior statements

Deck, 2005 WL 1200394, *5.

The Court turned next to the question of whether there is a constitutional difference when the shackling occurs not at the guilt-innocence phase, but instead at the sentencing stage of a capital trial. To decide that question, the Court reviewed the reasons that motivate the general prohibition against physical restraints. Although concerns about suffering may have originally been the basis for judicial hostility towards shackling, the Court’s more recent decisions have emphasized the importance of giving effect to three fundamental legal principles: (1) the presumption of innocence; (2) the right to counsel; and (3) the need for a dignified judicial process. Each of these legal principles can be jeopardized by shackling.

As for the permissibility of shackling at the sentencing phase of a capital trial, the Court concluded that the considerations it identified as disfavoring shackling at the guilt-innocence phase “apply with like force to penalty proceedings in capital cases.” 2005 WL 1200394, *8. The latter two principles identified above – securing a meaningful defense and upholding the dignity of the courtroom – clearly are implicated in the penalty phase of the trial. Although shackling at the sentencing proceedings does not effect any presumption of innocence, the Court nevertheless found that it “threaten[ed] related concerns.” Id. The Court explained: “Although the jury is no longer deciding between guilt or innocence, it is deciding between life and death. That decision, given the ‘severity’ and ‘finality’ of the sanction, is no less important than the decision about guilt.” Id, (citations omitted). Nor is the accuracy of the life or death decision any less critical. But the appearance of a defendant in shackles at the sentencing stage almost inevitably implies to the jurors that court authorities have determined that the defendant is a danger to the community. This is significant given that future dangerousness is nearly always a relevant factor in capital sentencing proceedings, even when it is not a statutory aggravating factor or even argued by the prosecutor as a reason for a death sentence. Furthermore, shackling almost invariably adversely impacts the jury’s view of the defendant’s character. Thus, physical restraints serve to undermine the jury’s ability to accurately weigh relevant considerations in making the sentencing decision, and can put a thumb on death’s side of the scale.

The Court therefore concluded that routine shackling of defendants is equally impermissible at the guilt-innocence and sentencing phases of a capital trial. In order to justify restraints, there must be “particular concerns, say special security needs or escape risk, related to the defendant on trial.” Id.

Finally, the Court addressed Missouri’s argument that the Missouri Supreme Court’s decision in Deck’s case met the constitutional requirements set forth above. Missouri contended that the state supreme court properly found: (1) the record did not show the shackles were viewed by the jury; (2) there was no abuse of discretion by the trial court; and (3) Deck suffered no prejudice.

As to what the jury saw, the Court found that Missouri was overstating the finding of the state supreme court. The record was clear that the shackles were visible; what was remarked upon by the state court was the absence of evidence in the record as to the extent to which the jury was exposed to the restraints. This goes only to prejudice.

The claim that the trial court properly exercised its discretion foundered on the absence of any indication in the record that the trial court was aware that the matter called for discretion. Notably, there were no findings, formal or informal. There was no mention by the trial court of a risk of escape or courtroom security. Further, although the trial court referred to the jurors’ fear of Deck, there was no explanation of why there was a special reason to be afraid of him. Nor was there an explanation for why the trial court permitted visible shackles at the sentencing phase after apparently ordering hidden ones for the guilt-innocence phase. The record simply did not demonstrate indisputably good reasons for the shackles.

Lastly, the question of prejudice was addressed. Missouri failed to take into account the Court’s statement in Holbrook that shackling is “inherently prejudicial.” Because the effects of shackles on the jury is unlikely to be shown from the trial transcript, a defendant is not required to demonstrate actual prejudice in order to make out a due process violation where a trial court allows a defendant to be seen in physical restraints without adequate justification. Instead, the burden is on the State to prove beyond a reasonable doubt that the shackling did not contribute to the death sentence.

The judgment of the Missouri Supreme Court was reversed and the case remanded for further proceedings not inconsistent with this decision.

Justice Thomas, joined by Justice Scalia, dissented. He argued that neither English common law nor the tradition of the States supported the majority’s conclusions about shackling. Thomas contended that the genesis of the common law rule against trying a man in irons was based on the intense physical pain experienced from the restraints which could impede the defendant’s ability to defend himself. The shackles that were used at that time are completely different from modern restraints, according to Thomas. Thus, the common law rule has no application as to modern shackles.

Thomas looked next to state practice to determine whether a deeply rooted tradition supported a finding that the Due Process Clause limits a trial court’s use of shackles. He found no such support, for three reasons: (1) state practice is of modern vintage, i.e., mostly 20th century; (2) the state cases have substantial differences; and (3) state and federal cases following Illinois v. Allen, Estelle v. Williams, and Holbrook v. Flynn, are not evidence of current consensus because they simply reflect dicta contained in the Court’s decisions.

Thomas further found that some of the traditional factors weighing in favor of shackling were present here, such as Deck’s assistance in an escape attempt by other prisoners, and the double murder he committed to avoid arrest. The majority’s conclusion that the record was insufficient to justify shackling is interpreted by Thomas as meaning that the flaw below was the absence of express findings by the trial court supporting the shackling order. Thomas found little sense in requiring on the record findings of what is obvious in the record.

Putting aside the question of shackling during a trial, Thomas contended that the requirement that a defendant be free from visible restraints during a capital sentencing proceeding has no foundation at all in either tradition or modern state practice. Further, in his view, it defies common sense. At sentencing, jurors are aware that the defendant has been convicted of a dangerous crime, and would know that the defendant is jailed at that point. In addition, jurors would understand that it makes sense to restrain the defendant at the sentencing phase where the defendant’s situation is the most dire. At that point, a defendant has little to lose should he attempt to escape and fail, and the courtroom provides the best opportunity to flee. Thomas also noted that a defendant at the sentencing stage may be angry, citing to cases where a defendant attacked his attorney, a witness and court reporter. Given what the jurors would understand about the situation, Thomas is unable to believe that a defendant would be prejudiced by appearing at sentencing in handcuffs.

As for the “modern” rationales for the majority’s rule, Thomas found they did not warrant a rule precluding shackling at sentencing. While accuracy in sentencing is a legitimate concern, because Thomas does not believe shackling at the sentencing stage is prejudicial he sees no accuracy concerns implicated by the shackles. Turning to interference with the ability to assist in one’s defense, there was nothing in the record showing any problems between Deck and his attorney because of the restraints. Nor was there any hint that the shackles were the reason why Deck chose not to take the stand and ask for mercy. Further, if shackles proved an impediment to testifying, they could be temporarily removed so the defendant could testify. Finally, Thomas accused the majority of misconstruing precedent with its invocation of courtroom decorum. No prior decision of the Court had even intimated that protection of the courtroom’s formal dignity was an individual right enforceable by a criminal defendant. Thomas was also unable to comprehend what was so offensive about seeing a murderer in shackles.

Thomas concluded his dissent by arguing that the majority’s purported concern for courtroom security rings hollow in light of the rule it adopted. He cited to numerous sources showing the courtroom to be a dangerous, or potentially dangerous, workplace, and ended with the statement: “The Court’s decision risks the lives of courtroom personnel, with little corresponding benefit to defendants. This is a risk that due process does not require.” 2005 WL 1200394, *20.

Medellin v. Dretke
544 U.S. ___, 125 S.Ct. 2088 (2005) (per curiam)

In a per curiam opinion, the Supreme Court dismissed Medellín’s certiorari petition as improvidently granted. Two questions had been presented to the Court in the case: (1) whether a federal court is bound by the International Court of Justice’s (ICJ) ruling that United States courts had to reconsider claims for relief under the Vienna Convention on Consular Relations raised by a number of Mexican nationals, including, José Medellín, without regard to procedural default doctrines; and (2) whether a federal court should give effect, as a matter of judicial comity and uniform treaty interpretation, to the ICJ's judgment. Following the grant of certiorari, President Bush had issued a memorandum stating that the United States would discharge its international obligations under the ICJ judgment by “having State courts give effect to the [ICJ] decision in accordance with general principles of comity in cases filed by the 51 Mexican nationals addressed in that decision.” Relying on that memorandum, Medellín filed a successor habeas petition in the Texas Court of Criminal Appeals and asked the Supreme Court to stay this case pending the state court’s ruling. Given that the Texas proceeding could provide Medellín with the very reconsideration of his Vienna Convention claim that he sought in federal court, and that the briefing before the Court revealed a number of hurdles Medellín must surmount before qualifying for federal habeas relief, the Court decided instead to dismiss the writ as improvidently granted. Justice Ginsburg wrote a concurring opinion, which was joined in part by Justice Scalia. Justice O’Connor dissented, joined by Justices Stevens, Souter and Breyer. Justice Souter authored a separate dissent, as did Justice Breyer, who was joined by Justice Stevens.

Medellín, a Mexican national, was sentenced to death in Texas. After his conviction and sentence were affirmed on direct appeal, he filed a state habeas petition claiming for the first time that Texas failed to notify him of his right to consular access as required by the Vienna Convention. The state court rejected the claim, finding that the Vienna Convention did not create individual, judicially enforceable rights and that the claim was procedurally defaulted. Medellín next filed a federal habeas petition which included the Vienna Convention claim. The petition was denied by the district court. While Medellín’s request for a certificate of appealability (COA) on the claim was pending in the Fifth Circuit the ICJ issued its decision concerning Medellín and other Mexican nationals facing the death penalty in the United States. The ICJ ruled that the Vienna Convention guaranteed individually enforceable rights and that the United States had violated those rights. The ICJ held that the United States was required to provide, by means of its own choosing, review and reconsideration of the convictions and sentences of the Mexican nationals. The review was to determine whether the violations “caused actual prejudice.” The ICJ further ruled that procedural default rules could not bar such review.

The Fifth Circuit denied Medellín's COA application as to the Vienna Convention claim, relying on Medellín's procedural default of the claim as well as circuit precedent holding that the Vienna Convention did not create an individually enforceable right. The appeals court did acknowledge the existence of the ICJ's judgment, but gave it no dispositive effect. Under existing Supreme Court precedent, Breard v. Greene, 523 U.S. 371 (1998) (per curiam), procedural default rules could be applied to a Vienna Convention claim. Certiorari was granted by the Supreme Court to address the questions listed above.

In dismissing the writ, the Supreme Court provided two reasons for its action. First, Medellín might receive the relief he wants from the state court. Second, certiorari review would be available from the state court decision, thus avoiding the thorny issues presented by the case in its present posture. The Court explained: “There are several threshold issues that could independently preclude federal habeas relief for Medellín, and thus render advisory or academic our consideration of the questions presented. These issues are not free from doubt.” The issues identified by the Court were: (1) whether under Reed v. Farley, 512 U.S. 339 (1994), a violation of the Vienna Convention is noncognizable in a federal habeas proceeding; (2) whether the deferential standard of 28 U.S.C. § 2254(d) would preclude relief; (3) whether Teague v. Lane would preclude reliance on the ICJ decision; (4) whether a COA can issue on a claim of a treaty violation; and (5) whether Medellín must show further exhaustion of state remedies in order to gain relief from the ICJ judgment or the President’s memorandum. Given that Medellín may receive what he seeks from the state court, the Court found it “unwise to reach and resolve the multiple hindrances to dispositive answers to the questions here presented.”

Justice Ginsburg authored a concurring opinion. She would have granted Medellín’s request for a stay, but because a majority of the Court did not support that action she believed that dismissal was the proper course. In the portion of her concurrence which was joined by Justice Scalia, Ginsburg emphasized that by dismissing the pending certiorari petition, the Court “would have jurisdiction to review the final judgment in the Texas proceedings, and at that time, to rule definitively on ‘the Nation’s obligation under the judgment of the ICJ if that should prove necessary.’” (Quoting Justice Souter’s dissenting opinion). Ginsburg justified the dismissal by reciting the “bewildering array of questions” that would remain unresolved by the primary dissent’s proposal to simply reverse the denial of the COA and return the case to the Fifth Circuit to address those issues in the first instance. This entire array of questions could be eliminated by the action of the Texas courts.

Furthermore, Ginsburg found flaw in the primary dissent’s determination that the Fifth Circuit should be required to compare the ICJ’s ruling with its own interpretation of the Vienna Convention. This, according to Ginsburg, is at odds with both the President’s order to “give effect” to the ICJ decision, and with the law of judgments. While the primary dissent maintained that this independent review of Medellín’s Vienna Convention claim is required by the second certiorari question, Ginsburg argued that Medellín “chose not to break out for discrete review in this Court questions underlying and subsumed in the ICJ’s judgments [at issue].” Medellín’s plea for international comity and uniform treaty interpretation does not, in Ginsburg’s view, seek a federal court’s independent interpretation of the Convention.

Also troubling to Ginsburg is the primary dissent’s failure to provide clear instructions to the Fifth Circuit “on which of the several questions the dissenters would remit to that court comes first, which others ‘should be part of’ the COA determination, and which are meet for adjudication only if, as, or when a COA is granted.” In addition, any participation of a federal court in “the fray” at this juncture risks a collision with the state court proceedings. Apparently recognizing this concern, the principal dissent noted that the Fifth Circuit could hold the case in abeyance until the Texas courts act. But Ginsburg failed to understand the justification of parking the case in the Fifth Circuit after refusing to put it on hold in the Supreme Court. Given that there are two new bases for relief presented to the state court, i.e., the ICJ judgment and President Bush’s memorandum, Ginsburg found that the Texas courts were now positioned to immediately adjudicate “these cleanly presented issues in the first instance.” The Supreme Court will retain the responsibility to provide ultimate answers if need be.

Justice O’Connor, joined by Justices Stevens, Souter and Breyer, dissented. O’Connor found that reasonable jurists would find debatable or wrong the district court’s disposition of the Vienna Convention claim. Therefore, she concluded the proper resolution was reversing the Fifth Circuit and remanding for consideration of three questions: (1) whether the ICJ judgment is binding on U.S. courts; (2) whether the Convention creates a judicially enforceable individual right; and (3) whether the Convention sometimes requires state procedural default rules be set aside so that the treaty can be given “full effect.”

O’Connor agreed with the Court that the AEDPA is applicable to Medellín’s Vienna Convention claim because it was enacted after the treaty was signed. Under the AEPDA, a COA can only issue if the applicant has made a substantial showing of the denial of a “constitutional right.” In O’Connor’s view, Texas had waived its right to argue that a COA cannot issue on a treaty-based claim. This was because the argument was made for the first time in its merits brief. Although Texas argued that the COA requirement is jurisdictional, and hence unwaivable, O’Connor was unconvinced. She found it to be “jurisdictional” in the sense that it is a gateway device, but not a true jurisdictional requirement which a court must raise and decide even without prompting by the parties.

O’Connor also was unpersuaded by Texas’s argument that § 2254(d)(1) is fatal to the treaty claim. Under that statute, if a claim is “adjudicated on the merits” by a state court, federal habeas relief is available only if such adjudication “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court.” Notably, when the Texas court applied its procedural default rule “it did not adjudicate the merits of the relevant federal question – whether, under Article 36(2) [of the treaty], the treaty overrides state procedural default rules.” In addition, the Texas court appeared to have reasoned that private individuals can never enforce a treaty in federal court. This blanket ruling was contrary to Supreme Court precedent. Because Texas never asked nor answered the correct question – whether an individual can bring a claim under the Vienna Convention – the federal court is to proceed de novo on that issue, unencumbered by § 2254(d).

As for the other nonjurisdictional issues that the Court thinks might preclude habeas relief for Medellín, O’Connor did not believe their existence justified dismissing the case. Instead, they were issues that should be resolved in the first instance by the Fifth Circuit. Regarding the potential applicability of Reed v. Farley, “a complicated question,” neither Texas nor the United States had mentioned the case thereby depriving Medellín of the opportunity to even brief the issue.

The second potential impediment noted by the Court was § 2254(d). The Court observed that Texas had arguably adjudicated the claim on the merits when it stated that Medellín had failed to show harm from Texas’s failure to notify the Mexican consulate of Medellín’s arrest. O’Connor pointed out, however, that while Texas had asserted in its certiorari papers that Medellín had received a prejudice analysis from the state habeas court, it abandoned that position in its merits brief. Texas instead argued that Medellín would be unable to show prejudice at some future hearing. Medellín, in turn, had consistently maintained that he had never received an evidentiary hearing on the question of prejudice. Further, from its order the ICJ appeared to agree with Medellín that no prejudice analysis had ever been conducted.

The third potential impediment to relief identified by the Court was Teague v. Lane. O’Connor noted that this was never discussed by the Fifth Circuit and concluded that “[w]hether Teague bars relief for Medellín is itself a highly debatable question that should be part of a proper COA analysis upon remand.”

O’Connor then provided a substantial assessment of Medellín’s arguments to demonstrate why, in her view, the COA’s denial should be vacated. In so doing, she noted a suggestion by the Court that Medellín’s reliance on the ICJ decision may constitute a distinct claim that is not exhausted. O’Connor countered that Medellín had consistently maintained a single claim – that Texas violated his rights under the Vienna Convention and that he is entitled to a remedy. She believed that Medellín’s reliance on the ICJ opinion “is akin to pointing to a new decision from this Court to bolster an existent claim for relief. In neither case has petitioner made a new claim as opposed to a new argument supporting his pending claim.”

In the process of discussing the second question presented, O’Connor expressed disagreement with Ginsburg about its scope. O’Connor accused Ginsburg of giving “an unduly narrow construction to the second question presented.” O’Connor construed the second question – whether a United States court should give effect to ICJ judgments for the sake of comity or uniform treaty interpretation – as requiring the domestic court to independently interpret the treaty, compare its interpretation with that of the ICJ, and then decide how and to what extent the two views should be reconciled.

Given the pending state proceedings premised on the Presidential memorandum, O’Connor did note that had her course of action been adopted, the Fifth Circuit may have wanted to stay its own proceedings until the successor petition was resolved by the Texas Court of Criminal Appeals.

Justice Souter also wrote a dissent. Souter believed that the best course, in light of the pending state habeas petition, would be to stay the certiorari proceedings. By doing so, the Court would “remain in a position to address promptly the Nation’s obligation under the judgment of the ICJ if that should prove necessary.” Because that position failed to garner the support of the majority, Souter concluded the next best course would be to “take up the questions on which certiorari was granted, to the extent of their bearing on the conclusion of the Court of Appeals that there was no room for reasonable disagreement, meriting a COA, about Medellín’s right to relief under the Convention.”

Souter observed that the Fifth Circuit felt itself constrained by Breard v. Greene, 523 U.S. 371 (1998) (per curiam), where the Court applied a procedural bar to a Vienna Convention claim. Given the existence of the new ICJ judgment “that may be entitled to considerable weight, if not preclusive effect,” the Fifth Circuit should be free to “take a fresh look” irrespective of Breard. Although Souter joined O’Connor’s dissent, it was with two caveats. First, he did not believe the Court should limit the scope of the proceedings on remand. The question whether or not the Fifth Circuit should construe the treaty independent of the ICJ’s judgment was not presently before the Court. Therefore, in Souter’s view, O’Connor’s discussion of the issue should not limit the Fifth Circuit’s inquiry if a remand were possible. Second, Souter would order the Fifth Circuit to stay the case pending the state court’s resolution of the pending habeas petition.

Also writing a dissenting opinion was Justice Breyer, joined by Justice Stevens. Breyer agreed with Souter and Ginsburg that a stay of the certiorari proceedings was the appropriate course. In the absence of majority support for that course of action, Breyer agreed with O’Connor’s approach. He agreed that the Fifth Circuit erred in finding that Medellin’s legal argument that American courts are bound by the ICJ judgment was not substantial. Breyer further disputed Ginsburg’s contention that a remand to the Fifth Circuit would invite rival proceedings, given the Fifth Circuit’s likely recognition that the unique circumstances favor a stay while the state proceedings run their course.

Bradshaw v. Stumpf
545 U.S. ___, 125 S.Ct. 2398 (2005)

In this Ohio capital case, the Court unanimously reversed the Sixth Circuit’s grant of relief on two of petitioner’s claims, and remanded for further proceedings on a portion of one of those claims.